Study Guide

Objective

The Student

- has an basic understanding of AI and the history of AI.
- has knowledge of where AI is used and its development today.
- has an basic understanding of different machine learning algorithms and their future possibilities
- can recognize different inputs used in AI and machine learning
- recognises the possibilities to get information and data from different systems and how Human - Machine Interaction is adapted in autonomous vessels

Content

- Administrative matters
- Computational thinking and algoritms - What is computing? Algorithms and complexity
- Introduction to AI and Autonomy - What is AI? How do you define Autonomy?
- Agents and Search - How to solve problems with “Good old-fashioned AI”
- Introduction to Machine Learning - Overview of ML, Risks and Problems with ML
- Supervised learning - Basic supervised learning through regression
- Machine vision - Deep neural networks, Image segmentation, Image detection, Image recognition
- Reinforcement learning - Reinforcement learning as search, Autonomy and reinforcement learning
- Industrial Internet - What is IoT? What is a Digital Twin?
- Sensors and Sensor fusion - Situational awareness, LIDAR, IR, GNSS and IMU’s
- Autonomy and Safety - Software safety, Liability, Accountability

Evaluation scale

H-5

Assessment criteria, satisfactory (1)

Sufficient 1
Theory and methodology are poorly understood and implemented in Autonomous Maritime Operation related tasks/ assignments.
Research, communication and documentation are hardly acceptable.
Active participation.
Satisfactory 2
Appear to grasp theory and have made a start in showing its applicability in Autonomous Maritime Operation related tasks/ assignments. Research, communication and documentation are acceptable.
Active participation.

Assessment criteria, good (3)

Good 3
Understanding of theory and applicability of methods in Autonomous Maritime Operation related tasks/ assignments, but work could be stronger.
Research, service design, communication and documentation are good.
Active participation.
Very Good 4
General understanding of theory and methods, very good implementation in Autonomous Maritime Operation related tasks/ assignments.
Reliable research, innovative service design and communication and documentation on good level.
Very active participation.

Assessment criteria, excellent (5)

Excellent 5
Mastery of theory and methods, proficiency of implementation of them in Autonomous Maritime Operation related tasks/ assignments.
Outstanding research, innovative service design and excellent communication and documentation.
Very active participation.

Qualifications

No prerequisites.

Objective

The Student

- has an basic understanding of AI and the history of AI.
- has knowledge of where AI is used and its development today.
- has an basic understanding of different machine learning algorithms and their future possibilities
- can recognize different inputs used in AI and machine learning
- recognises the possibilities to get information and data from different systems and how Human - Machine Interaction is adapted in autonomous vessels

Content

- Administrative matters
- Computational thinking and algoritms - What is computing? Algorithms and complexity
- Introduction to AI and Autonomy - What is AI? How do you define Autonomy?
- Agents and Search - How to solve problems with “Good old-fashioned AI”
- Introduction to Machine Learning - Overview of ML, Risks and Problems with ML
- Supervised learning - Basic supervised learning through regression
- Machine vision - Deep neural networks, Image segmentation, Image detection, Image recognition
- Reinforcement learning - Reinforcement learning as search, Autonomy and reinforcement learning
- Industrial Internet - What is IoT? What is a Digital Twin?
- Sensors and Sensor fusion - Situational awareness, LIDAR, IR, GNSS and IMU’s
- Autonomy and Safety - Software safety, Liability, Accountability

Materials

Lecture materials

The intelligent systems institute @ Novia collects useful resources and study material related to AI and machine learning in our public GitHub repository:

https://github.com/NoviaIntSysGroup/resources-and-learning-material/blob/main/Study_Material.md

Teaching methods

Teaching methods:

- Lectures,

- Assignmets (coding, presentations, and reports),

Exam schedules

No exam, grade is based on course assignments.

Further information

The intelligent systems institute @ Novia provides instructions for installing relevant software and for setting up your own computer to work with machine learning projects in our public GitHub repository.

https://github.com/NoviaIntSysGroup/resources-and-learning-material

Evaluation scale

H-5

Assessment criteria, satisfactory (1)

Sufficient 1
Theory and methodology are poorly understood and implemented in Autonomous Maritime Operation related tasks/ assignments.
Research, communication and documentation are hardly acceptable.
Active participation.
Satisfactory 2
Appear to grasp theory and have made a start in showing its applicability in Autonomous Maritime Operation related tasks/ assignments. Research, communication and documentation are acceptable.
Active participation.

Assessment criteria, good (3)

Good 3
Understanding of theory and applicability of methods in Autonomous Maritime Operation related tasks/ assignments, but work could be stronger.
Research, service design, communication and documentation are good.
Active participation.
Very Good 4
General understanding of theory and methods, very good implementation in Autonomous Maritime Operation related tasks/ assignments.
Reliable research, innovative service design and communication and documentation on good level.
Very active participation.

Assessment criteria, excellent (5)

Excellent 5
Mastery of theory and methods, proficiency of implementation of them in Autonomous Maritime Operation related tasks/ assignments.
Outstanding research, innovative service design and excellent communication and documentation.
Very active participation.

Assessment methods and criteria

Pls see 'Study Unit Information'.

Qualifications

No prerequisites.